distinct count of multiple columns

[shankyy7227]

My data contains FName, LName, MName, Gender, Card ID, Health ID, Active Flag and there may be Null in any column for each row i am trying to calculate distinct count  (FName+Card ID+Health ID) and distinct count  (FName+Card ID+Health ID+Where Gender=M)

FNAME	LNAME	MNAME	Gender	Card ID	Health ID	ActiveFlag
AB	CD	X	M	1234567	12345	Y
BC	EF	Y	M	9810561	67891	N
AB	CD	X	M	1234567	12345	Y
IH	PQ	Z	M	4567890	34564	N
IM	PY	C	F		56789	N
			F	1456478	67890	N
ME	EE	B	M		78901	N
AB	CD	X	M	1234567	12345	Y
BC	EF	Y	M	9810561	67891	N

Please help me with this.

Thanks

[Anonymous]

Hi,

I am wondering which is the best solution in terms of performance?

Concatenate Calculated Column ?
OR
GroupBy Measure ? 

Thank you in advance,

Rob

[matemusic]

I think for performance is better calculated column, for datamodel size is better groupBy.

Correct me someone, if I am wrong.

[marcorusso]

Yes, calculated column is better for performance.

The measure will not perform well in that version, it's better to use SUMMARIZE instead of GROUPBY and CALCULATETABLE instead of FILTER, this way you don't have to materialize Gender:

Measure2 = 
CALCULATETABEL (
    COUNTROWS (
        SUMMARIZE ( User, User[FNAME], User[Card ID], User[Health ID] )
    ),
    User[Gender]="M"
)

[jlopezfelizzola]

Maestro!

Just a quick precision:

Measure2 = 
CALCULATE (
    COUNTROWS (
        SUMMARIZE ( User, User[FNAME], User[Card ID], User[Health ID] )
    ),
    User[Gender]="M"
)

 Change the CALCULATETABLE to simple  CALCULATE.

Have a great day!

[Anonymous · Accepted solution]

Create two measures and try the following Dax

Measure1 = COUNTROWS(GROUPBY(User,User[FNAME],User[Card ID],User[Health ID],User[Gender]))

Measure2 = COUNTROWS(FILTER(GROUPBY(User,User[FNAME],User[Card ID],User[Health ID],User[Gender]),User[Gender]="M"))

[Anonymous]

Worked for me too.

Thumbs up

[shankyy7227]

Thanks Jessica the Measures are working.

[Anonymous]

Hi

1. Create a new column :   New_Col = 'Table1'[FNAME]&'Table1'[Card ID]&'Table1'[Health ID]

2. Then you can create measure to do the count as needed:

Dist_cnt = CALCULATE(DISTINCTCOUNT('Table1'[New_Col]))

Dist_Male_Cnt=CALCULATE(DISTINCTCOUNT('Table1'[New_Col]),'Table1'[Gender]="M")

Hope this helps.

Thanks
Raj